When It Would Be Money

It is fashionable to say that money is a shared hallucination. That line flatters our cleverness while excusing our passivity. Hallucinations cannot settle debts across adversarial jurisdictions; hallucinations do not finance supply chains. Money works because it is backed by a machine, sometimes a literal machine of war, sometimes a machine of law. This chapter makes the case for a third: verification machines that continuously produce scarce, indispensable utilities.

Post-Bretton-Woods, the machine was compliance. The machine this thesis builds is verification. Cypherpunks did not abolish trust; they automated it. When privacy, proofs, and compute clear across neutral rails, money stops asking for favors and starts paying for facts.

Stated as a conditional, the thesis reads:

If a dense digital civilization continues to rely on AI, global networks, and programmable markets, and if states continue to use repression and narrative control rather than explicit default, and if the full service path remains stress-deliverable, value accrues without bypass, and a self-custodied loss-bearing holder constituency persists, then a bearer base asset may earn monetary premium by preserving settlement, proof, compute access, and exit when ordinary substitutes weaken—provided project credit remains separate and without assuming the triad must share one token.

The argument, restated as a conclusion chain:

1. Digital civilization structurally needs privacy, proofs, and compute.

2. These capacities can be standardized and verified through the seven-layer stack.

3. Standardized work can produce cheap-to-check receipts.

4. Receipts represent capacity only where the full path delivers usable service under stress.

5. Stress-deliverable receipts can support markets.

6. Markets can produce recurring fee flows.

7. Fee flows support a scarce base asset only under non-bypassable accrual rules.

8. A persistent, self-custodied, loss-bearing, regime-responsive constituency must warehouse that asset’s risk.

9. Only then, if the asset remains liquid, neutral, verifiable, legally holdable, and agency-preserving, may it earn a store-of-value premium.

The asset may first behave like infrastructure equity, a claim on fee flows from useful services. Only later, if liquidity deepens, neutrality is sustained, and holdability matures across jurisdictions, does it potentially cross the threshold from “productive asset” to “monetary collateral.” That transition is not guaranteed; it is the thesis’s central conditional claim, and it has a base-rate problem the thesis should state precisely rather than bury or overstate. Stated precisely: no equity-like claim on a fee stream is known to have made this crossing. Warehouse receipts, bills of lading, standardized commodity claims, and the utility tokens of Complete Edition, Chapter 34: Objections & Responses^† (the closest historical analogs to Work Credits and fee units) acquired price and liquidity without ever acquiring monetary premium.

The restriction to fee-stream claims matters, because the broader statement (that no asset in monetary history has crossed from use-value to money) is false, and false against the account of commodity money this thesis in fact relies on. Menger [Menger 1892] argued that money arises without design when the most saleable good in an economy comes to be accepted for its saleability rather than its use. Mises [Mises 1912] closed the regress by requiring that a money’s purchasing power trace back to a prior exchange value it had as a commodity. Silver, salt, grain, colonial Virginia’s tobacco, and privately issued banknotes made that crossing. The path they took (use-value, then saleability, then general acceptance) is the path this thesis proposes for a base asset whose use-value is access to verified service, and Menger’s account is an argument for that path’s possibility, not against it. What the record does not contain is a case in which the object making the crossing was a claim on an operator’s future fees. Every such claim stayed a claim.

The thesis is therefore not describing a developmental stage other assets have passed through, and it is not describing something without precedent either: it is proposing that a bearer asset can take the Mengerian path while a fee stream is routed through it, which is the combination for which the base rate is empty. What follows is the argument for that possibility. Complete Edition, Chapter 30: Risk Analysis & Failure Modes^† supplies the piece Menger requires and this chapter cannot supply on its own: why the saleability bid should attach to the base asset rather than to the typed service claims it prices.

The Mengerian path has a machine-speed variant worth one paragraph. Saleability, in Menger’s account, is the property of being accepted without haggling over what the thing is. For a software agent transacting across a trust perimeter that property is not a convenience but a precondition: it cannot investigate a counterparty’s unit, so it can accept only units anyone can check (Chapter 6: The Delegation Economy: Who Buys Verification). Acceptance-without-inspection by agents is therefore a candidate first form of the saleability step for an asset whose unit quality is publicly verifiable. This is an argument that the path is open to that constituency, not evidence that anyone has taken it. It adds no precedent: no agent economy has made anything money, the base rate above is unchanged, and Red Line 17 read on agent borrower panels is where the claim would show up or fail to.

The media-transition literature offers a reading of the fee-stream failure record, and it is offered as a reading, not as evidence: a printed scroll is still a scroll. Each of those precedent systems delivered the content of a new capacity (storage, rendering, compute) while keeping the form of the old one: a service purchased from an operator, priced as a service, held as an operating claim. Content without structure is exactly what a medium looks like in its first generation [McLuhan 1964], and the utility-token record is that first generation’s ledger. On this reading the crossing the thesis proposes is not an increment on those systems but a change of kind: fees are the content, and the holder-side stress-deliverable service is the medium. That is why the base rate is empty, and why the empty base rate is not dispositive. It is also why the reading earns nothing: no link of Chapter 9: The Triad and the Monetary Candidate strengthens because a pattern from another century rhymes. The chain still has to prove itself, link by link, in telemetry. The honest summary of the two moves is this. The empty base rate is evidence against: it raises the prior that the crossing fails, as the five named precedents and the empty commodity-claim record should. The change-of-kind argument is the reason the evidence is not dispositive: it identifies a property the precedents lacked, which changes what the base rate is a base rate of. Neither move makes the crossing likely; together they make it testable, which is what the red lines are for.

As a Store of Value

From the SoV lens, a credible store of value must be credibly scarce, cheap to verify in public, resist censorship and capture, have native demand that is not purely narrative, and avoid being a duration instrument whose real return can be pinned negative by policy.

Privacy, Proofs, and Compute may support a base asset that meets that brief, but the services themselves and their Work Credits are not presumed monetary:

• Privacy is purchased because some people and institutions must pay without broadcasting their graph: dissidents, NGOs, treasuries under capital controls, enterprises with sensitive payroll and vendor relationships. In a repression-heavy world, privacy is not a luxury good; it is the hull that keeps savings from becoming an option owned by someone else.

• Proofs are purchased because “seeing is believing” has failed. Deepfakes, platform curation, and the liar’s dividend make any unproven artifact suspect. Regulated AI and finance regimes require auditable provenance and computation. In that world, proofs are not a niche; they are the affidavit layer of the digital order.

• Compute is purchased because intelligence is now a first-class input to production. FLOPs for training, inference, and proving are line items in budgets. Once those FLOPs are wrapped in proofs and standardized as canonical workloads, verified compute becomes a commodity that can be priced, hedged, and stored.

In a world that may choose stealth default (negative real yields aided by regulation) over explicit default, a durable SoV must be duration-neutral, peg-proof, stress-deliverable, and held by actors able to bear loss. Recurring utility can support value accrual; it cannot supply moneyness by itself.

The three services above are the use-value. The monetary question is why a holder who wants them under stress would hold the base asset rather than the typed Work Credits that promise them directly, alongside Bitcoin for exit. That question is answered once, at Complete Edition, Chapter 30: Risk Analysis & Failure Modes^†, and not restated here: Work Credits are claims on specific capacity that fails in the states the hedge is for, Bitcoin hedges exit and not access, and the base asset is the only workload-agnostic bearer instrument the protocol pays in. The same passage concedes the corollary: if holders in fact prefer Bitcoin plus Work Credits, or Bitcoin converted on the day, the access experiment and Red Lines 16 and 18 record it, and the thesis treats that as its default null.

Duration-neutrality of the claim is not, however, a theory of how plants get built. Complete Edition, Chapter 3: The World Forces New Monetary Primitives^† and Complete Edition, Chapter 34: Objections & Responses^† state the completeness condition the rest of this chapter does not repeat: reserve collateral and duration finance are different functions; proofs can audit the second; they must not become it. A stack that supplies Privacy, Proofs, and Compute still requires some institution, outside Layers 0–6, capable of warehousing the construction interval. That institution is not the token.

As Stack

Seen from the stack angle, the triad is backed not by a metaphor but by a supply chain:

• Layer 0 keeps machines honest and powered (verifiable hardware, FERs).

• Layer 1 keeps packets flowing under censorship (VerifyReach).

• Layer 2 keeps code moving even when app stores and CDNs are hostile.

• Layer 3 keeps identity accountable without doxxing.

• Layers 4–5 make work and value flow through proofs and privacy rails.

• Layer 6 keeps governance and telemetry legible.

Triad instruments are claims with different seniority and risk. Work Credits represent typed future access or service capacity; FCR, FER, and PIDL artifacts provide evidence; project notes bear duration; LP and staking positions are derivatives or operating claims. Only the base asset is tested as a monetary candidate.

As Telemetry

Seen from the telemetry angle, every link of the chain has an instrument, and each instrument asks a different question. VerifyPrice asks whether anyone can still check the work. VerifyReach and VerifySettle ask whether checked work can still be reached and paid for under pressure. Delivered Verified Capacity asks whether the service survives the narrowest edge of its physical path (Complete Edition, Chapter 17: Layer 0: Verifiable Machines & Energy^†). VerifyFlow asks whether the price is telling the protocol’s story or the wrapper’s (Complete Edition, Chapter 26: Extended Telemetry^†). Beneath them sit the holder instruments, the Native Monetary Buyer Map and the Principal Map, which ask what none of the others can: whether anyone holds the asset through stress, and whether a human still stands behind the holding (Complete Edition, Chapter 26: Extended Telemetry^†).

Together they are the constitution. They keep “trustlessness” from decaying into “trust the custodians,” and they keep “store of value” from decaying into “the price went up.”

A system that cannot show its own health cannot be money; it can only be marketing.

Trading One Base Reality for Another

What changes and what stays the same?

What changes: The base reality that backs money. Instead of “the sovereign will repay” or “gold is scarce because geology,” the base reality becomes “these capacities are scarce, necessary, and verifiable.”

What stays the same: Money is still a claim on work. The work just becomes specific and measurable: proofs that anyone can check, privacy that anyone can use, compute that anyone can verify.

The operative word is agency. Monetary arrangements can be read as answers to the question: who has agency over their own savings, payments, and economic identity? Gold answers it with physical possession. Fiat answers it by delegation to institutions. Bitcoin answers it with self-custody and cryptographic keys, and answers it well enough that this thesis builds its settlement layer on top of the answer rather than around it. The triad addresses adjacent surfaces that self-custody does not reach: agency over verification (proofs), over private settlement (paying without being watched), and over productive capacity (compute without permission). Different surfaces, not a later stage (Chapter 7: Why Not Something Already Here?). A store-of-value premium, where it exists, accrues to assets that preserve agency on whichever surface is under attack.

The case for the possibility is made; what remains is where a premium would actually come from. The answer is three mechanisms on the chain’s last link: pledgeability without due diligence, the state-contingent holder-side flow, and denomination. Each is argued here and read on its own red line (Chapter 17: Risk Analysis & Failure Modes).

Pledgeability Without Due Diligence

Nobody examines a banknote. That is not carelessness; it is the property that makes it money. A thing accepted only after inspection is being traded; a thing accepted without inspection is being spent.

The distance between those two is what the study of collateral and short-term funding means by moneyness. Not scarcity. Not cash flow. Not convenience yield. An asset is money-like when no counterparty gains anything by investigating the particular unit tendered: nobody looks, so it passes at face value, and a thing that passes at face value can be pledged without due diligence [Gorton & Pennacchi 1990] [Dang et al. 2020] [Holmström 2015]. Deposits, bills, and repo collateral circulate on that property. A funding run is not a change of mind about fundamentals. It is the moment looking starts to pay.

Two kinds of information.

Transparency alone produces nothing of the kind, since equities are the most transparent securities in existence and among the most information-sensitive. The claim holds for one kind of information only. Fundamental information is what the asset will be worth: cheap verification does nothing to it, the thesis claims nothing about it, and a triad asset stays as sensitive to it as gold does. Unit-quality information is whether this unit, this receipt, this pledged position is what it purports to be.

Why did a market full of experts stop accepting collateral at face value in 2008? Not because anyone changed their view of housing; every participant had held one for years. What nobody could answer cheaply was whether a particular pool was good, so lenders began producing private information about specific collateral, haircuts widened, and no-questions-asked acceptance ended [Gorton & Metrick 2012]. Moneyness failed on unit quality while fundamentals were already common knowledge. The same split says where collateral demand would lean if private views on fundamentals grew more valuable: toward public unit quality, the one axis on which nobody can hold an edge.

Information-Insensitivity by Symmetric Knowledge

There are two routes to no-questions-asked acceptance, and they are opposites.

Symmetric ignorance is the historical route. Producing information about the specific collateral is expensive relative to the gain, so nobody produces it, and the asset trades at par. Opacity is the mechanism, not a defect of it.

Symmetric knowledge is the route a verification-native asset takes. Producing that information costs approximately nothing and the result is public, so no counterparty can come to hold an informational edge over another. The asset trades at par for the opposite reason.

Both deliver acceptance without due diligence. They differ in how they end. Symmetric ignorance is a knife-edge: it holds only while nobody looks, and the first agent to look acquires an advantage that every other agent must then defend against, which is the run. Symmetric knowledge has no such trigger, because there is no private information about unit quality available to produce; private views about fundamentals remain, and are priced as volatility rather than as due diligence. You cannot start a run by looking when looking is free and public, because looking buys nothing that everyone else could not have had for the same nothing.

Deterrence, not audit.

The box does not claim that anyone looks. Universal audit is observed nowhere, even where looking is close to free: public filings sit unread, and the record is routinely complete and unanalyzed [Green 2026f]. An argument that needed every counterparty to verify would be refuted on its first day, and would deserve to be.

This argument needs nobody to look: what removes the run trigger is the absence of a discoverable edge, and free public verification removes that whether or not anyone exercises it. The agent who looks acquires nothing the agent who declines could not have had at the same cost. A security camera does most of its work on the days nobody watches the tape. The discipline falls on the claimant, before the claim is made, the way the credible possibility of entry disciplines a contestable market without entry being observed [Baumol 1982]: a misrepresentation anyone could falsify at zero cost is not worth attempting, so it is not attempted, so nobody has to check. The equilibrium described here is one in which verification is constantly available and almost never used, and those two facts are the same fact.

Two consequences follow. What to measure is the cost of verification, not its observed rate: Red Line 1 reads the standing threat, and falling verification activity is not evidence against the mechanism while the cost holds. And a system can be perfectly auditable and monetarily inert, because deterrence works only if being caught would cost the claimant something, a condition that Red Line 14 and Chapter 13: Why Hold the Base Asset? instrument rather than assume.

The position, not the token.

That a unit of a public ledger is authentic is true of every cryptocurrency, and nobody has ever had to investigate whether a bitcoin is a bitcoin. That is table stakes, and it has never made a token money.

The differentiator is the collateralized position. Pledge the asset against an obligation to deliver a service (a prover bond, corridor collateral, SLA escrow) and a lender has to assess the bundle: whether the capacity behind the obligation exists, whether the operator’s slashing exposure is what it claims, whether the physical substrate delivers under the scenario the haircut assumes. That is where due diligence lives, and it is what Facility Capacity Receipts, energy receipts, Delivered Verified Capacity (Complete Edition, Chapter 17: Layer 0: Verifiable Machines & Energy^†) and the grade ladder make cheap, which is why Layer 0 is a monetary requirement rather than an engineering preference (Chapter 4: First Principles: What a SoV Must Survive).

The thesis already prices this. Gold, expensive to smelt and cheap to assay, is information-insensitive on unit quality, which is why a stranger’s nugget could be accepted without investigating the stranger; every haircut keyed to Layer 0 grade, sovereign optionality, or bridge class (Complete Edition, Chapter 17: Layer 0: Verifiable Machines & Energy^†, Complete Edition, Chapter 25: Layer 6: Governance & Telemetry^†) is a price on residual unit-quality uncertainty. The document derived an information-sensitivity schedule from engineering intuition without the monetary theory that explains what it was pricing.

Necessary, not sufficient.

Graded commodities are cheaply assayed and warehouse receipts never became money. Certified diamonds are authenticated to a published standard and are not money. Bearer bonds are authenticable at sight and are not money. Cheap unit verification is therefore not sufficient, and any argument that stops here proves far too much.

What separates the cases is a conjunction. Insensitivity on unit quality becomes monetary only alongside fungibility (certification grades a stone without making stones interchangeable), durability and portability (grain is gradeable, perishable, and bulky), and the absence of a counterparty (a warehouse receipt brings back the custodian, and a bearer bond is somebody’s liability). Gold is the one historical object at that intersection: assayable, fungible, durable, portable, and nobody’s promise. The claim available here is narrow, and it is the whole of what this section asserts:

A verification-native digital bearer asset is the second class of object to occupy that intersection—and the first whose assay cost does not scale with the size of the claim being assayed.

That last clause is the only new term. Assaying more metal costs more; verifying a succinct proof costs the same whether the claim beneath it is worth a dollar or a billion. If information-insensitivity is what makes collateral money-like, an assay invariant to notional is a change in kind rather than in degree.

The forger gets better tools too.

That is the strongest objection to any cheap-verification argument: if fabricating plausible false claims gets cheaper as fast as checking them, nothing has changed. In natural language the objection is fatal, because telling true from false degrades as the adversary invests. Cryptographic verification has no such gradient. Under stated soundness assumptions (the hardness and parameter assumptions, and a circuit and verifier that constrain what they claim) a false proof is not expensive but infeasible, and rejecting one costs what accepting a valid one costs. A thousand forged proofs are a thousand rejections, because nothing outside the proof is consulted. Verification cost is invariant to the size of the claim and invariant to the effort of the adversary, and the second invariance is the one that survives contact with an opponent who is also getting better tools.

The bound is as firm as the claim (Complete Edition, Chapter 10: Proofs as Attestation Money^†): a proof is a bounded attestation, not truth, and says nothing about whether the workload was worth running or whether the sentence a human attached to the artifact is honest. The invariance holds only over claims that can be stated as a computation, a class much narrower than the one on which trust is failing, and outside it the thesis has nothing to offer.

Free riding, and discretion.

Grossman and Stiglitz [Grossman & Stiglitz 1980] showed that information nobody is paid to produce does not get produced. Proof systems escape the trap because cost and benefit fall on different parties: the prover pays to be believed, and every counterparty consumes the result at negligible cost. Verification is a private good bought by the claimant, and its public checkability cannot be withheld without destroying what was bought. Read from the financing side, competition prices proving toward its cost (Complete Edition, Chapter 22: Layer 4: Truth & Work^†), which is why the fee stream is ordinary service revenue (Chapter 13: Why Hold the Base Asset?). The property that makes verification sustainably supplied is the same property that stops it from being a source of excess return, and the thesis should not have expected to keep both. Whatever premium exists is not in the proving business, which is why Chapter 15: The Three Monetary Mechanisms and Chapter 15: The Three Monetary Mechanisms look for it somewhere a valuation cannot follow.

Issuance discipline, Condition 4 of the Value Capture Lemma, belongs here, and not as scarcity. A supply set by discretion is information-sensitive on supply: every counterparty must form and maintain a private view of what the issuing body will do, which is due diligence by another name. Credible non-discretion is not only a scarcity property. It is the condition under which the supply schedule stops being something anyone has to have an opinion about.

How it fails.

Symmetric knowledge reverts to ignorance when verification stops being affordable, concentrates until most participants trust a verifier instead of verifying, or loses its receipt data. Each reads as a degradation of service, and each is also a monetary run trigger. A fourth mode has no red line: scope drift. A position legible on its proof and opaque on its operator’s off-ledger leverage, counterparty concentration, or rehypothecation is 2008 with better cryptography. It is published as the off-ledger exposure disclosure series (Complete Edition, Chapter 26: Extended Telemetry^†), which feeds the Red Line 17 panel and the Red Line 10 opacity reading. The gap is a number, not a caveat.

What it predicts.

A reader of Dang, Gorton and Holmström [Dang et al. 2020] will object that insensitivity belongs to the payoff: drive information cost to zero on a state-dependent payoff and it becomes more sensitive. The objection is correct, and it fixes the scope. The argument here is adverse selection on unit quality, not a claim about payoffs. Bitcoin is the control: its units have been freely verifiable since 2009, and its haircuts sit at 30–50% and vary widely across lenders, because they price realized volatility and liquidation depth, which no verifier compresses. The thesis does not predict lower haircut levels.

It predicts something narrower. Quoted by the same lenders at the same moment, against opaque collateral of similar volatility, a position whose backing is publicly verifiable should show lower cross-lender dispersion, and dispersion that does not widen relative to the comparables when the market does. The flatter the payoff, the more of the prediction applies: an over-collateralized, DVC-backed position, debt-like over the states the haircut assumes, carries all of it, and the bare base asset a weaker share, since a lender who wants a flat payoff builds one from the asset and a haircut. A funding run’s signature is dispersion widening before level, as the better-informed lenders leave first; if that signature appears here, against the comparables and with receipts available, symmetric knowledge did not hold.

Is it measured? Every quarter, against a pre-committed panel of lenders: haircut level, cross-lender dispersion as the primary series, the cost of onboarding the asset at a new venue, and dispersion under stress. Is it sized? No. Red Line 17 reads the prediction as its box specifies (Chapter 17: Risk Analysis & Failure Modes), on a design that lets Bitcoin’s native units, set against their wrapped representations, pass or fail rather than fail by construction. Nothing in it says who keeps the gain: the saving goes first to the borrower who posts the position, and whether it travels on to the asset’s holders is read from the buyer map’s sell-through and retained balances (Complete Edition, Chapter 26: Extended Telemetry^†), a reading with no trigger of its own.

The section claims no price stability, no insensitivity on fundamentals, and no automatic acceptance by any lender, and it raises no rating in Chapter 9: The Triad and the Monetary Candidate. It supplies a mechanism and an instrument for link 9, not evidence that link 9 holds. Ratings that rise because an argument was added are self-congratulation.

Acceptance and demand.

Information-insensitivity is why a counterparty would accept the asset without investigating it; the holder-side flow of Chapter 15: The Three Monetary Mechanisms is why a holder would want it when other arrangements fail. Acceptance without demand is a payment rail; demand without acceptance is a collectible. Chapter 15: The Three Monetary Mechanisms asks the third question, whether anyone will reckon in it. The thesis argues all three, sizes none, and publishes an instrument for each.

Where a Premium Would Come From

If fees, burns, and collateral yield a well-valued cash-flow claim rather than money, the monetary claim has to be relocated, not abandoned. It relocates to something the thesis has been describing all along without calling it the monetary argument.

$\Delta$ depends on the regime.

Write $\Delta(R)$, with $R$ the regime-pressure index of Stage A (Chapter 16: Stages, Phases and Macro States). $\Delta$ is thin in benign states because a solvent, suable indemnitor substitutes for a proof (Chapter 13: Why Hold the Base Asset?). Turn that sentence over and it says when $\Delta$ is large: when the counterparty cannot be sued, will not be solvent when it matters, is itself the adversary, or inflicts a loss no damages restore (censorship, seizure, deplatforming, disclosure). That list is the threat model of Chapter 5: Threat Model, restated as a demand curve.

$\Delta$ is small in benign states and large in the tail. The asset’s usefulness is state-contingent, and its distribution is skewed toward exactly the states everyone else’s arrangements stop working in.

The tempting next step is to say that a payoff arriving when wealth is scarce must be worth more than its discounted expectation. It fails the test already set: a monetary premium is value in excess of discounted cash flows (Chapter 13: Why Hold the Base Asset?), and a valuation run with the correct stochastic discount factor already weights every cash flow by its state price. So the argument splits in two.

A better utility.

Because $\Delta(R)$ rises when the substitutes fail, the fee stream is countercyclical, and an investor should pay more for it per unit of expected cash flow. That is a discount-rate effect, fully representable inside a discounted-cash-flow valuation. A countercyclical utility is a better utility, not a monetary asset. Conceded, and set aside: nothing below rests on it.

A service to the bearer.

What the holder obtains is not a cash flow, so there is nothing to discount. It accrues to whoever holds the unit, by holding it, and a valuation of distributions omits it however its discount rate is set (one that prices services from possession, as transactional-demand models of tokens do, need not). It is why a bearer instrument held outside the banking system, with no cash flows at all, can carry a price. Storage theory has the mechanism as convenience yield ([Kaldor 1939]; [Working 1949]), the holder of inventory owning an option against stockout that the holder of a delivery claim lacks; the analogy is to the mechanism only, since a triad asset is not a storage commodity and takes no magnitude from that literature.

A fire extinguisher and a contract for the delivery of one both have a price. Only one works during the fire, and every argument here about regime states is an argument about when the market is on fire. An extinguisher is not money, though. It establishes a benefit of ownership; the step to monetary use is independent acceptance and a persistent demand to hold, which this section does not supply (Chapter 15: The Three Monetary Mechanisms argues the acceptance).

Gold makes the same point from the other side: no cash flows, negative carry, a storage-theory yield near zero in normal regimes, and a price every constructible DCF falls well below. On the standard account the residual is monetary and portfolio demand, the holder-side category reached from another direction.

Regime-Contingent Convenience Yield

The monetary premium available to a triad asset is the holder-side service flow generated by the state-contingency of $\Delta(R)$ — the bearer’s ability to transact, prove, hold, and exit when the substitutes for proofs, courts, and custodians have stopped working — conditioned on credible non-discretion in issuance and on bearer holdability. Those two are not added terms but enabling conditions: the flow is only a monetary service while issuance discipline holds and the unit remains bearer-holdable, and it collapses if either fails. The service accrues by virtue of holding rather than as a distribution, so there is no stream to discount. That is what places it outside a discounted-cash-flow valuation rather than merely at a different discount rate inside one.

It is not the countercyclicality of the fee stream. That is a negative beta on cash flows: real, valuable, and captured exactly by a DCF using the correct stochastic discount factor, which is why it confers no moneyness.

It is also not generated by burns, by collateral, or by fee share. Those produce a cash-flow claim and a collateral lockup (Chapter 13: Why Hold the Base Asset?).

What it rests on.

A premium carried by the holder-side flow rests on whether a holder can use the thing under pressure, and none of that is fee routing. Of the five lemma conditions, only issuance discipline touches the channel, because credible non-discretion is part of the payoff rather than of the accrual. Link 8 asks for a persistent self-custodied constituency able to bear loss. The premium itself hangs off link 9: liquidity, neutrality, verifiability, legal holdability, and agency preservation (Chapter 4: First Principles: What a SoV Must Survive).

A burn operates on the claim, and the premium is not on the claim.

Now discount it honestly.

The payoff is negatively correlated with the states the holder fears, which is the point; the delivery is positively correlated with them. $\Delta(R)$ spikes under energy interdiction, hardware denial, network filtering, and jurisdictional pressure (Complete Edition, Chapter 5: Threat Model^†, Complete Edition, Chapter 17: Layer 0: Verifiable Machines & Energy^†), the states that impair the protocol’s ability to deliver at all. Red Line 13 reads that as a threat to the hinge; it is also a discount on the hedging premium. The holder is asked to pay for insurance whose underwriter is exposed to the insured event.

The slots are occupied, too. Gold already holds “works when the network does not” and Bitcoin “works when the state does not”. A triad asset needs power, reachable networks, and reference hardware to deliver anything, so it hedges a narrower band: conditions adversarial enough to need proofs and privacy, not severe enough to take the grid. That band may still be the important one: repression that stops short of collapse is what Complete Edition, Part I: Context & Claims^† argues is most likely. Sovereign optionality (Complete Edition, Chapter 17: Layer 0: Verifiable Machines & Energy^†) exists to widen it, and disruption-adjusted VerifyPrice (Complete Edition, Chapter 17: Layer 0: Verifiable Machines & Energy^†) to price whether it has.

Unsized.

The mechanism is ordinary: service flows to a bearer are common in asset pricing. The holder-side flow is nonetheless unsized. Nothing in the thesis estimates the convenience yield a triad asset could command, and no honest estimate exists before there is a market to observe one in. That refusal is why the chain-strength rating for the store-of-value premium is Medium (conditional), and why the claim sits on link 9 rather than on fee accrual. The question is meant to be settled, not asserted: Complete Edition, Chapter 26: Extended Telemetry^† publishes lending-rate, basis, and wrapper-basis series against the regime-pressure index, and a series flat across regime states is the null result, reported as such.

Denomination

Ask a merchant in Buenos Aires what a thing costs and the answer comes in dollars, whatever ends up crossing the counter. That is the unit-of-account question, the one monetary function the other instruments do not measure: not whether anyone holds the asset or pays with it, but whether anyone chooses to quote in it.

It is also the stickiest function. A store of value can be abandoned in an afternoon: the holder sells and is out. A unit of account is welded into contracts whose counterparties would all have to move at once, each eating the basis if it moves alone, which makes it a coordination device, hard to acquire and correspondingly hard to lose [Doepke & Schneider 2017].

The trap, first.

A naive denomination share reads one hundred per cent on launch day and means nothing, because every triad service is priced in the base token by design (Chapter 9: The Triad and the Monetary Candidate) and Condition 1 of the Value Capture Lemma requires it. Mandated denomination is a routing rule. Everything of interest is in the complement.

Free-Choice Denomination Share

Let $\mathcal{C}_{\text{free}}$ be the set of contracts in which the parties were not required by protocol rule to express price terms in the base asset — forward and term capacity purchases, SLA agreements above the mandated fee, bilateral prover–buyer contracts, corridor quotes, over-the-counter blocks, service retainers, employment and vendor agreements among operators, and any obligation denominated between counterparties rather than by the protocol.

Free-choice denomination share is the fraction of $\mathcal{C}_{\text{free}}$ whose price terms are written in the base asset, reported by contract count, by notional, by tenor bucket, and by counterparty class.

The mandated complement — fees, required collateral postings, protocol-set tier pricing — is published separately and is never added to it. It is a design commitment and reads as one.

Quoting is not settling.

A contract quoted in dollars and settled in the asset is a payment-rail observation, and it is the configuration essentially all crypto commerce has taken. Bitcoin is the most successful new store of value of the last century on the measures this thesis uses, and after seventeen years it denominates almost nothing. Either denomination is not necessary for moneyness, or bitcoin’s moneyness is incomplete along a measurable axis. The thesis takes the second reading and concedes it is contestable.

The reading rests on one asymmetry. What bitcoin buys is produced by an economy that prices in dollars, so a bitcoin quote imports basis risk against every input the seller pays for. In a triad protocol the unit and the goods come from one system: a prover paid in the unit, posting collateral in it and buying capacity priced in it holds a naturally two-sided position, exposed to basis only on the legs that reach outside. That is a coordination path to denomination a pure bearer asset lacks. It does not say the path will be taken.

The walled-garden test.

Airline miles and in-game currencies denominate large internal economies and are not money, so what matters is where the counterparty stands. Tier I contracts have a protocol service on both legs, Tier II on one, Tier III on neither. Monetary content rises from I to III, the tiers are never blended, and Tier I is context rather than evidence: a protocol scoring highly there alone has built a company scrip with excellent telemetry. At the boundary the share is read against the dollar list price the buyer already has for compute.

The shape it predicts.

Nothing prices in a volatile unit if it can avoid it, so denomination should appear first at short tenor and first in two-sided positions, where the variance nets, and last, or never, in long-dated one-sided contracts. A share rising uniformly across buckets reads as subsidy or mandate. The cost that makes volatility prohibitive is the cost of carrying the exposure, which is what Chapter 15: The Three Monetary Mechanisms lowers: information-insensitivity lowers the carrying cost of denomination.

Confirming, not leading.

When populations abandon a domestic unit, savings dollarize years before prices do. If that ordering holds here, free-choice denomination confirms monetary adoption on the macro path rather than warning of it. It leads only on the protocol’s own contract layer, where Tier I and Tier II contracts can be counted from Phase I onward. What is early is the measurement, not the phenomenon.

Instrument and falsifier.

The share is published quarterly on the Value Capture Board (Complete Edition, Chapter 26: Extended Telemetry^†) on the custody and reproducibility rules Red Line 14 sets, so the party collecting the fee may not classify the contracts, beside a quote/settle matrix, a tenor and sidedness split, a subsidy disclosure (a purchased quote is not a chosen one) and redenomination at renewal. Red Line 18 reads the result in two conditions, outsiders declining to reckon in the unit and participants declining on unmandated contracts, each with the consequence its box sets (Chapter 17: Risk Analysis & Failure Modes).

The section does not claim denomination is necessary for moneyness, or likely; it raises no rating in Chapter 9: The Triad and the Monetary Candidate and promises no early warning on the macro path. It answers the question the other instruments cannot: what would be visible at Phase I that distinguishes this from every service token that ever shipped? A contract nobody was required to write in the unit, written in the unit anyway, with a counterparty outside the perimeter, at a tenor long enough to matter, without a rebate.

The mechanisms say where a premium would come from. This chapter says how it would arrive, if it does: the combined thesis as separable stages from need to price, the negative case of useful infrastructure without money, the phases of adoption, and the policy states they unfold in.

From Need to Price

A Three-Stage Model

The combined thesis can now be stated as three linked but separable processes.

Stage A: regime pressure creates structural need. Let $R_t$ index financial repression, surveillance, synthetic media, and AI concentration. Compute demand is deliberately not a component of $R_t$: it is a demand driver in its own right, entering alongside regime pressure rather than as part of the pressure index, so that $D_t^{\text{triad}} = g(R_t)$ is not partially defined in terms of its own output. Regime pressure generates demand for triad capacity:

$$D_t^{\text{triad}} = g(R_t)$$

This is the causal layer Complete Edition, Part I: Context & Claims^† develops.

Stage B: protocol design converts need into native value. Let $\phi_t \in [0,1]$ measure the effectiveness of the monetary design: required fees, burn share, collateral, issuance discipline, and non-bypassability. Then native captured demand is

$$D_t^{\text{native}} = \phi_t \, D_t^{\text{triad}}$$

As $\phi_t \rightarrow 0$, the system can remain enormously useful while the monetary object fails. This is not the Value Capture Lemma restated as a coefficient: the lemma is an AND-gate on five conditions, while $\phi_t$ is an empirical capture fraction those conditions constrain. The lemma says partial capture is possible only when the gate is open; $\phi_t$ measures how much of the triad demand actually flows through once it is.

Stage C: market structure realizes native value as a price path. Observed price depends not only on native demand but on aggregate exposure flows and the liquidity available to absorb them:

$$\Delta \ln P_t = \theta_t + I(Q_t, \Lambda_t) + \eta_t$$

where $\theta_t$ is the non-flow fundamental or native monetary drift (defined to exclude flow effects, so that all flow impact enters once, through the $I(Q_t, \Lambda_t)$ term, rather than twice), $Q_t$ is aggregate exposure demand (a flow over the interval $[t-1, t]$), $\Lambda_t$ is the liquidity and market-impact state, and $\eta_t$ is residual. Aggregate demand decomposes by source:

$$Q_t = Q_t^{\text{native}} + Q_t^{\text{allocation}} + Q_t^{\text{levered}} + Q_t^{\text{dealer}} + Q_t^{\text{trend}}$$

Only the first term is monetary evidence. The remaining four can dominate the price for extended periods, in either direction. Complete Edition, Appendix H: Formal Model of Market Realization, Wrapper Flows, and Price Capture^† develops the flow mechanics formally (holder flow elasticity, the net mechanical gain coefficient, the recycling boundary, market impact, and volatility drag), and Complete Edition, Chapter 26: Extended Telemetry^† defines the telemetry that makes the plane observable.

Why a protocol should care about external market structure. A designer might object that this is someone else’s problem. It is not. External wrappers affect treasury behavior, collateral demand, governance concentration, liquidity-provider incentives, public understanding, security-budget expectations, regulatory pressure, and the price that the protocol’s own participants use when making decisions. The protocol should not manage its price. It must nonetheless instrument the structures through which its monetary claims are represented, for the same reason it publishes VerifyPrice: unmeasured dependencies are where theses go to die quietly.

Stage B is where the monetary object can fail while the system stays useful, and that failure has a recognizable shape.

The Negative Case: Useful Infrastructure Without Money

If the triad remains useful infrastructure but never becomes money, it looks like this:

• Proof markets exist and grow, but fees are paid in stablecoins or fiat. The native asset has weak demand.

• Privacy corridors work, but most users route through custodial services that accept traditional payment.

• Verified compute is valued, but hyperscalers dominate and price in USD. Decentralized alternatives remain niche.

• The base token trades as a speculative asset correlated with crypto markets, not as a store of value with independent demand.

• Fee coverage never reaches 30%; the security budget depends on issuance subsidies indefinitely.

In this scenario, the stack succeeds (it delivers real privacy, proofs, and compute) but the monetary thesis fails (the native asset does not capture the economics). This is the “utility-token trap” outcome. It is not catastrophic (the infrastructure is still valuable), but it means the store-of-value claim was wrong.

The stack can win without the coin ever becoming money.

The telemetry makes this outcome visible: fee coverage, native-asset fee share, bypass-channel indicators, and holder behavior all signal whether the monetary thesis is tracking or not.

This scenario is not hypothetical. It maps directly to one pole of a live debate among Bitcoin practitioners: the position that Bitcoin succeeds as an escape hatch and self-custody tool for a small elite, but does not disrupt the existing power structure or impose a free market. The thesis’s value-capture conditions and telemetry regime exist precisely to make this distinction measurable rather than a matter of temperament.

Which outcome arrives, the stack with money or the stack without it, is read along the adoption curve.

Four Phases of Adoption

The curve runs as four overlapping phases, each pulled in by a different constituency: cypherpunks, then allocators, then developers who stop noticing the stack, then states. They vary by geography and sector, and they are metric-anchored, not calendar-anchored: a phase has arrived when its telemetry shows it, not when a date does. Calendar horizons are omitted on purpose, since they age badly and invite cheap dunking. Each phase has an exit gate, the reading that opens the next, and a failure reading, the one that sends it back or lands on a red line.

Phase I: cypherpunk-led.

The people who already live in the future (open-source cryptographers, hardware hackers, privacy-chain communities, early AI and ZK builders) prove that the loop of create, prove, settle, and verify works without permission. Their artifacts are reference implementations and receipts: open-admission prover markets publishing live VerifyPrice dashboards, testnets that measure rather than assume whether verification overhead stays inside its band on commodity hardware, public receipt ledgers, and BTC$\leftrightarrow$XMR swaps that people actually use, the one non-custodial privacy corridor deployed today. Most users are still speculators and hobbyists. A few things nonetheless become facts rather than hopes: proofs can be bought from permissionless markets, paid for over privacy rails without custody, and checked by anyone with a laptop.

Exit: VerifyPrice for $\geq 3$ canonical workloads with $\geq 6$ months of history and p95 inside its SLO under stress; VerifySettle for $\geq 1$ corridor at $\geq 95\%$ success, with no protocol-attributable refund failure and overall refund safety $\geq 99.5\%$, every failure root-caused; VerifyReach published for major regions with real multi-path reachability; and $\geq 1000$ receipts a day. The question then stops being whether this can exist and becomes whether it can scale. Failure: verification cost above its SLO for three consecutive months without credible remediation is Red Line 1’s reading, a refund failure the protocol caused on an admissible route is Red Line 2’s, verification concentrated $>70\%$ on one profile or jurisdiction for three consecutive months is Red Line 3’s once Phase II has opened, and receipt data going dark leaves the thesis unverifiable.

Phase II: allocator-led.

Fund managers, corporate treasuries, exchanges, and large platforms start to treat privacy and proofs the way they once treated bandwidth: as operating costs, not science projects. They reclassify. A bank’s AI risk model runs on verifiable compute because regulators ask for receipts; a media platform attaches provenance because the liability of not doing so is too high; a trading venue settles privately rather than leak its order book. None of them has converted to cypherpunk ideology. Compliance and risk management now require math, not memos. Budgets stop paying for promises and start paying for work that anyone can check.

Exit: fee coverage $\geq 30\%$ of the security budget for 12–24 months, counted as retained fees with burns excluded; budgeted workloads outnumbering speculative ones in the receipt tags; VerifySettle inside its SLOs through at least one policy shock; capex and opex lines that name verification; and, where agents originate material demand, Principal Map readings inside their pre-declared bounds (Complete Edition, Chapter 26: Extended Telemetry^†). The 30% sits roughly an order of magnitude above anything the closest precedents reached (Complete Edition, Chapter 34: Objections & Responses^†). It is set where the security budget stops being issuance-funded, a deliberately hard target with a known failure rate.

Allocators can arrive two ways that look alike on a price chart and nothing alike in the telemetry. Native adoption buys, stakes, and posts the base asset, pays native fees, and holds self-custodied claims; it moves every series on the Value Capture Board. Wrapper-led exposure buys funds, treasury-company equity, and futures; it may precede native adoption, seed it, or permanently replace it, and only VerifyFlow says which (Complete Edition, Chapter 26: Extended Telemetry^†). Demand from software agents reads as native on the boards; whether it is monetary adoption depends on whose principal issues the delegations, since a fleet’s demand is exposure by another route (Chapter 6: The Delegation Economy: Who Buys Verification). Assets under management, market capitalization, price, and wrapper launches therefore do not qualify the system for Phase III.

False Phase Transition: Wrapper-Led False Positive

Market capitalization and institutional AUM rise rapidly while native use, fee coverage, burns, and collateral remain flat. The ecosystem appears to have advanced to institutional adoption but remains, economically, in the speculative phase. This is the most flattering way for the thesis to fail, which is exactly why it needs a named gate rather than a footnote.

Failure: fee coverage back below 30% with a workload mix $>80\%$ speculative returns the system to Phase I, and a deeper, longer collapse is what Red Line 5 reads. The fee base is also late by nature. Budgeted verified compute is disproportionately a deployment-phase demand, one that plausibly becomes measurable only on the far side of a financial clearing event the thesis does not predict and cannot time (the capital-cycle clock of Chapter 16: Stages, Phases and Macro States), so through a disorderly period the fee base should look worse before it looks better. That is a re-timing, not a relaxation: the threshold and the regression gate are unchanged, and Condition B of Red Line 14 reads the opposite case, a clearing event that came and went with no fee base behind it. Phase III is not reachable on a 36-month horizon, and every deliverable that can be dated is a Phase I artifact.

Agents can also invert the order. They need no treasury committee, only a receipt schema, a refund-safe corridor, and a unit anyone can check, which is Phase III’s defining property shipped by construction. No gate is relaxed for that, and composability that is one operator’s SDK is a fleet, not a phase.

Phase III: composability-led.

The stack disappears into infrastructure, the way TLS did on the web. New applications emit claims and receive receipts because it is easier than hand-rolling trust, frameworks ship verification modules, and payments that cross borders route quietly over privacy rails with receipts that satisfy auditors but not censors. From the outside nothing dramatic happens; what changes is the default. Who carries that default to the median user, a carrier’s channel or the user’s own agent, is read as the Cross-Perimeter Share at the consumer edge (Chapter 6: The Delegation Economy: Who Buys Verification), a series not yet computable. This phase can show durable cash-flow accrual to the asset. Whether a monetary premium sits on top is a separate question, decided by the holder-side flow (Chapter 15: The Three Monetary Mechanisms) and not answered by price in either direction (Chapter 13: Why Hold the Base Asset?).

Exit: the base asset as collateral or reserve in $\geq 3$ independent native venues, with the Native Monetary Buyer Map showing a self-custodied, loss-bearing cohort rather than inferring one from lockup; VerifyPrice and VerifySettle stable through macro stress; a cohort holding the base asset for years, with Work Credit inventory reported separately as service demand; and fee coverage $\geq 50\%$, sustained. The base asset has then advanced as a monetary candidate, and Work Credits have not changed instrument class. Failure: the Value Capture Board reads Phase III on one operator’s traffic.

False Phase Transition: Fleet-Led Composability

Receipt volume, native fee share, and collateral lockup all rise because one operator’s agent framework routes through the protocol by default; Cross-Perimeter Share is near zero or not computable and the Human-Revocable Share is undefined. The Value Capture Board reads Phase III; the Principal Map reads Phase I with a large customer. Publish the second reading.

Phase IV: policy-led.

Once Phase III holds through macro stress, the curve turns political. Some jurisdictions codify cryptographic receipts as evidence, mandate provenance for certain media and AI systems, and recognize privacy rails with viewing keys as compliant infrastructure. Others keep their chokepoints. If the rails stay non-custodial and censorship-resistant, and proof and compute markets stay globally addressable, capital and talent can route around them the way data routed around telcos. Real once: receipts codified in $\geq 2$ jurisdictions, regulation that references proof primitives rather than records, public institutions publishing their own receipt and corridor metrics, and volume migrating measurably toward jurisdictions that respect lawful privacy. In the best case each wave of repression pushes savings onto rails where verification is cheap, and each wave of adoption deepens those markets. That is a best case, not a forecast, and the premium in it remains the unsized quantity of Chapter 15: The Three Monetary Mechanisms.

Any phase can fail backward. A slide from Phase II back to Phase I is a failure of adoption, not of the technology, and the telemetry exists so the difference can be seen. Nor is the curve guaranteed. Verification cost can creep, hardware or router oligopolies can re-centralize control, regulation can pinch the ramps harder than expected. If the stack stays anchored in verifiable machines, keeps verification cheap and public, treats privacy as infrastructure and forces its own claims through telemetry and receipts, this is the curve it can plausibly aim at.

Macro Policy States

AfterFiat is a structural architecture, not a tactical Treasury short. The same sovereign system can pass through different states, and policy repair can extend incumbent arrangements for years, so the model names states and observables rather than a dated sequence. Its market-repair distinction and repair sequence are adapted from Michael Green’s commentary [Green 2026c] [Green 2026d], which supplies mechanisms and scenarios, not adopted quantities or tactical forecasts.

State 1: market repair succeeds.
Buybacks, maturity changes, regulatory repair, or lower front-end rates restore market function while long-dated inflation compensation stays contained. The immediate repression premium for an open bearer stack may narrow even while the secular agency case remains.

State 2: physical inflation blocks repair.
Energy, power, shipping, or industrial bottlenecks lift long-dated inflation compensation. The financial system may need easier policy while the physical system argues for restraint, and Layer 0 costs rise just as demand for neutral assets may. Long-dated breakevens and inflation swaps are the tripwire.

State 3: repair fails and repression follows.
No durable marginal buyer returns, and policy shifts toward captive demand, preferential regulation, negative real returns, capital controls, or restricted exit.

State 4: competent closed-stack stabilization.
The state coordinates energy, industry, compute, identity, payments, and duration warehousing; functionality and collateral stability improve while user custody and exit contract. This is not collapse. It is the open stack’s strongest competitor.

Transitions are conditional and read from observables, and none of them supplies a tactical forecast for bonds, currencies, or commodities. Three readings cut across all four states. Household transmission: a state can be stable at the sovereign level while its households deplete, and a State 1 with exhausted buffers moves toward whatever regime the depleted median will vote for (Chapter 3: The World Forces New Monetary Primitives). The capital cycle is a clock rather than a state [Perez 2002], borrowed for one mechanism and not for its historical periodization, which the thesis does not adopt: installation finance outruns what the economy can absorb, a Turning Point reprices the installed base, and a Turning Point arriving in State 2 meets a central bank with less room to absorb it than one arriving in State 1; the phase gates of Chapter 16: Stages, Phases and Macro States are read against it. Duration absorption, not the debt ratio, decides whether repair holds, because the marginal buyer is increasingly demand commissioned by benchmark rules and financed in repo (Chapter 3: The World Forces New Monetary Primitives). None of this times the transition; it says the plumbing series would show it first (Complete Edition, Appendix I: Scenario Analysis: The Collateral Loop Under Stress^†), and the long yield is not among them.

Recession first, repression later is a path, not a promise.